I have spent more hours than I can count sitting in clinic waiting rooms in mainland China. Not as a patient. As the person responsible for making a digital transformation programme actually land across a private healthcare chain that was opening sites faster than anyone could standardise them.
The thing I remember most is not a system or a dashboard. It is a paper appointment book. A large one, with a plastic cover, sitting on the front desk next to a terminal that ran perfectly good software. The receptionist used both. She wrote the appointment in the book first, then entered it into the system later, when things were quiet. Sometimes much later. Sometimes not at all.
If you had looked at the product usage data, you would have concluded there was a feature gap. Booking is clunky. We need a faster flow, a better calendar view, maybe a mobile app for front desk staff. That is the diagnosis almost every founder reaches, and it is almost always wrong.
The book was not there because the software was slow. The book was there because when the owner walked in at seven in the evening and asked how the day had gone, he looked at the book. The book was the system of record for the one person whose opinion mattered. The software was a thing head office cared about. Until you change what the owner looks at, you can ship a hundred releases and the book stays on the desk.
The pattern: you are not losing to a competitor, you are losing to paper
When a Western product struggles in Asia, the Gulf, Southeast Asia or Latin America, the founder’s instinct is to look inward. We must be missing something. Local language support. A payment method. An integration. Some compliance box. So the roadmap fills up with market-entry features, the engineering team spends two quarters on them, and the conversion rate moves by nothing worth measuring.
I have made this mistake myself. When we were building one of the early location-based platforms in China, we spent real energy on product sophistication at a moment when the market had not yet decided it wanted to behave digitally at all. We grew past three hundred thousand users, but almost none of that growth came from the clever parts of the product. It came from being physically present where the behaviour we wanted already happened, and from people telling other people. The product was a reason to stay. It was never the reason to start.
The same lesson arrived again in hardware. Building consumer products in China with the Jean-Michel Jarre venture, we took one product and grew the line to eight. The engineering was demanding and the supply chain was unforgiving, but the hard part was never whether the thing worked. It was whether the channel wanted to carry it, whether the person on the shop floor could explain it in thirty seconds, and whether the buyer had an existing mental slot to put it in. A product with no slot is a product that gets explained, and anything that needs explaining does not scale through a channel you do not control.
So here is the pattern, stated plainly. In emerging markets your competitor is not the other vendor in your category. Your competitor is a process that currently works. Paper works. WhatsApp works. The owner’s memory works. A cousin who handles the accounts works. These systems are free at the point of use, they have zero onboarding cost, and they have survived for years. Nobody wakes up wanting to replace something that works.
Why the obvious approach fails
Four things break at once, and they break in ways that look like product problems.
The buyer’s workflow is analog, so your software is not replacing a system, it is creating one. That means every hour your customer spends with your product is a net new hour, not a saved hour. The ROI story you built for a market where the baseline is another piece of software collapses here, because the baseline is a notebook that costs nothing.
The decision-maker is the owner, not a committee. In a mature market you sell to an IT function, a procurement process, a security review. Frustrating, slow, but legible. In a lot of emerging markets, the clinic owner, the distributor, the family that owns the four sites, decides in one conversation. There is no evaluation framework for you to win. There is only trust, and trust is not acquired by a free trial.
The purchase is relationship-collateralised rather than ROI-justified. I mean something specific by this. The buyer is not underwriting your software, they are underwriting the person who introduced it. If that person disappears, the contract is dead regardless of usage. Which is why churn in these markets so often correlates with your account manager leaving, not with your feature set.
And nobody will pay to change a process that currently works. This is the one founders refuse to accept. They assume inefficiency creates demand. It does not. Inefficiency creates demand only when someone inside the business is personally punished for it. Find that person or do not bother.
The standard response to all of this is to hire a growth marketer, run paid acquisition, build a self-serve funnel, and wait for product-led growth to do what it did at home. In a market where the buyer does not yet behave digitally, product-led growth is a machine with no fuel. You are pouring budget into a top of funnel that leads to a behaviour nobody has.
The operator’s alternative
Your first hire is not a growth marketer. It is a local operator who will sit in the waiting room.
I am being literal. The most valuable commercial asset I have ever had in Asia was a person who could spend a morning in a client’s office, watch what actually happened, drink tea, be told the real reason something was not being used, and come back with information that no survey and no analytics tool would ever surface. The paper appointment book did not show up in a product review. It showed up because somebody was physically there.
This person needs three qualities. They must be credible to the owner, which usually means they come from the industry rather than from software. They must be comfortable doing unglamorous work, because the first year is unglamorous. And they must have enough authority to change the commercial terms on the spot, because deals in these markets close in the room or not at all.
Second, anchor your pricing to a cost the buyer already pays in cash. Not to the value you create, not to a percentage of savings, not to a seat-based model imported from a market with different labour economics. Find the line item that already leaves their bank account every month. A part-time administrator. A commission to an agent. A fee to the person who does the reconciliations. Price against that, visibly, and the conversation becomes a swap rather than an addition. An addition needs justification. A swap needs only to be slightly better.
Third, accept that your onboarding is a service business for the first hundred customers. You will do the data migration yourself. You will sit with the receptionist for two days. You will configure the account, train the staff, come back in three weeks because they reverted, and train them again. Founders hate this because it does not look like software economics. It is not supposed to. You are buying behaviour change with your own hands because there is no other currency that works at this stage. The margin comes later, once you know exactly which fifteen steps of that process can be automated, and you only learn that by doing all fifty of them manually first.
Distribution is the product, the software catches up
The founders I have watched win in these markets did something that looks irrational from a Western product perspective. They built the route to the customer before they built the thing the customer was buying.
They found the person who already visits two hundred clinics a quarter, or the association everyone belongs to, or the equipment supplier whose technician is in the building anyway, and they made themselves useful to that channel first. The software followed. Sometimes the software was embarrassing for the first eighteen months. It did not matter, because distribution compounds and features do not.
In banking innovation work in Hong Kong, building machine learning into lending decisions and hyper-personalisation, the modelling was the fun part and the smallest part. What determined whether anything reached a customer was whether it fitted into a distribution reality that already existed. The best model in the building is worth nothing if it cannot be delivered through the channel the bank actually has.
So if your expansion into Asia or the Middle East is stalling, stop opening the roadmap. Go and count. How many times has someone from your company been physically inside a customer’s place of business this quarter? If the answer is under twenty, you do not have a product problem. You have an absence problem.
Spend the next quarter on three things. Put one credible local operator in the field with the authority to close. Rewrite your pricing so it displaces a cash cost the buyer already recognises. And commit, out loud, to doing the onboarding by hand for your first hundred accounts.
The book will still be on the desk for a while. That is fine. You are not trying to remove it on day one. You are trying to make the owner look at your screen instead, just once, at seven in the evening. Everything else follows from that.
I write from twenty years of building businesses between Europe and Asia. If your company is facing this, start a conversation.